Next, an optimized TensorRT engine is built based on the input model, target GPU platform, ... Our example loads the model in ONNX format i. lite.
確定! 回上一頁